Many inorganics are known or suspected human carcinogens, such as arsenic, cadmium, and lead. These agents pose significant hazards to the population of the United States after environmental, occupational or intentional exposures. Defining carcinogenic mechanisms greatly aids in designing prevention or intervention strategies and in assigning appropriate levels of risk to these exposures. The primary goal of the ICS is to define the molecular mechanisms of carcinogenic action of arsenic, cadmium, and lead under the project titled Molecular Mechanisms of Inorganic Carcinogenesis. Since they can impact carcinogenicity and potentially provide a means for prevention or intervention, mechanisms of transport and acquired tolerance are studied as well. Emphasis is also placed on factors that may dictate heightened susceptibility to inorganics, such as early-life exposures and poor expression of critical adaptive genes. These inorganics attack various targets in humans. Arsenic causes skin, urinary bladder, lung, kidney, liver, and prostatic malignancies and possibly uterine cancers. Cadmium has been primarily associated with human lung, prostatic and kidney cancers and possibly pancreatic cancers. Lead exposure has been linked to kidney and brain malignancies. Thus, various in vitro and in vivo model systems have been developed to study important molecular targets and targets tissues, with an emphasis on human relevance. These inorganic carcinogens likely have multiple mechanisms that are site and cell specific, and our focus has been on epigentic factors. Significant advances and future directions include: &#9679; A reproducible rodent model where inorganic arsenic acts a complete carcinogen has been developed in which brief in utero arsenic exposure in mice leads to tumors or proliferative lesions of the urogenital system, liver, lung and adrenal in the offspring as adults. The urogenital system lesions include transplacental arsenic-induced or initiated tumors of the ovaries, uterus, vagina, and bladder and proliferative lesions of the kidney. These results are in accord with human studies that indicate the liver, urinary bladder, lung, kidney and uterus are target tissues of arsenic carcinogenesis. Molecular mechanism studies indicate disruption of estrogen signaling by in utero exposure to arsenic contributes to the liver, lung and urogenital system malignancies, in part through aberrant activation of estrogen receptor-&#945;. Indeed, we find that tumors and proliferative lesions of the urogenital system, including the uterus, ovary, vagina and urinary bladder, are greatly enhanced by postnatal exposure to synthetic estrogens like diethylstilbestrol. We also find evidence of aberrant estrogen signaling in arsenic exposed human liver. Further molecular characterization of arsenic-induced in utero tumor initiation is underway, including aberrant gene imprinting, using this model of arsenic carcinogenesis. We hypothesize that arsenic in utero may attack a critical pool of progenitor cells in target organs and induces aberrant genetic reprogramingas part of its carcinogenic mechanism. These studies have important public health implications, including the identification of fetal period as a time of very high sensitivity to arsenic and the possibility that co-exposure to pharmacological or environmental estrogens could enhance development of arsenic-initiated cancer. Further study will include prenatal arsenic exposure combined with exposures to urinary bladder and renal tumor promoters in mice to enhance the carcinogenic response to arsenic in these key human target organs. &#9679; Various in vitro cell transformation model systems have also been developed to study inorganic carcinogenesis. In doing these studies we select cells with relevance to the human targets of arsenic, cadmium or lead carcinogenesis, and use low-level exposures for long periods, which approximates typical human exposures and avoids supra-physiological responses associated with acute high doses that could have limited relevance to the carcinogenic process. A human prostate epithelial cell line has been malignantly transformed with cadmium and arsenic, both potential human prostatic carcinogens. Additional work indicates the arsenic and cadmium transformants both acquire androgen independence, an event associated with a very poor clinical prognosis in prostatic cancer patients, largely through androgen receptor by-pass related mechanisms. Molecular dissection of the events associated with arsenic- or cadmium-induced malignant transformation in this and other human cell lines will continue with a focus of aberrant expression of genes critical to the carcinogenic process. In addition, a human prostate progenitor cell has been developed and will be studied as a potential target cell population of these carcinogenic inorganics. Furthermore, we have successfully transformed a human pancreatic ductal cell with cadmium, which fortifies a possible role of cadmium in this deadly disease. Similarly, arsenic has induced malignant transformation of human skin keratinocytes. The study of this arsenic-induced skin cancer model indicates that it occurs through a very different mechanism from internal cancers, one which involves apoptotic by-pass and aberrant survival of damaged skin cells. Transformation experiments of human renal cells with cadmium and lead are underway. Finally, we are developing a data base that provides unique patterns of gene expression and altered DNA methylation that may allow us to molecularly define sub-sets of prostate, pancreatic, skin, or renal cancers in humans as being attributable to arsenic, cadmium and/or lead exposures with tumor tissue obtained through the cooperative human tissue network. &#9679; Lead exposure causes inclusion body formation, and although these aggresomes contain precipitated protein and the majority of the lead within the cell, they are otherwise poorly characterized. Lead-induced inclusions appear to protect critical cellular targets by sequestering large amounts of this poorly excreted metal. Metallothionein (MT) is a metal-binding protein that normally binds zinc but also detoxicates toxic metals. We find that MT-knockout (MT-null) mice are hypersensitive to lead toxicity, including nephrocarcinogenicity, because they cannot make lead inclusion bodies. Similarly lead exposed MT-null cells do not form inclusions in vitro. MT appears on the outer surface of inclusions in lead-exposed MT-wild type (MT-WT) mice. Many diseases show similar aggresomes, like Parkinsons and Alzheimers, which are tentatively linked with lead exposure. Precipitated &#945;-synuclein (SN), a chaperone protein, is found in these neurodegenerative inclusions. In fact, we find MT-null cells poorly express SN but transfection of MT into MT-null cells restores expression. Co-precipitation experiments indicate MT may directly bind to SN. Studies are underway to test the hypothesis that SN may assist formation of lead-induced aggresomes. These studies could have important implications for lead carcinogenesis, zinc physiology and neurodegenerative diseases potentially linked to lead. Furthermore, MT levels in humans vary widely for unknown reasons and poor MT production may be a key predisposing factor to lead toxicity and carcinogenesis. Studies are underway to further test this hypothesis. For instance, our work in MT competent mice shows transplacental lead exposure induces a modest level of kidney tumor formation in the offspring as adults, so we are now studying the nephrocarcinogenicity of transplacental lead in MT-null mice to see if this response is enhanced. Because there are clearly human populations particularly sensitive to lead, any factor that predisposes individuals to lead toxicity takes on great importance